vol. 169, no. 4
the american naturalist
april 2007
Adaptive Advantages of Cooperative Courtship for Subordinate
Male Lance-Tailed Manakins
Emily H. DuVal*
Department of Integrative Biology and Museum of Vertebrate
Zoology, University of California, Berkeley, California 94720
Submitted July 18, 2006; Accepted November 27, 2006;
Electronically published February 6, 2007
abstract: Male lance-tailed manakins (Chiroxiphia lanceolata) cooperate in complex courtship displays, but the dominant (alpha)
partner monopolizes mating opportunities. This raises the question
of why subordinates (betas) cooperate. Three nonexclusive hypotheses explain the adaptive basis of helping behavior by subordinate
males: cooperation may increase (1) subordinates’ immediate reproductive success, (2) the reproductive success of close relatives, or (3)
subordinates’ chances of future reproduction. I demonstrated that
beta males rarely sired chicks and were unrelated to their alpha
partners but received delayed direct benefits from cooperation; betas
had an increased probability of becoming an alpha when compared
to males that had not been betas. To investigate the mechanism by
which betas attain these adaptive benefits, I examined betas’ success
in replacing their alpha partners both in natural turnover events and
when alphas were experimentally removed. Beta males did not consistently inherit alpha roles in the same territories where they served
their beta tenure, arguing that queuing for status does not fully
explain the benefits of cooperation for betas. Instead, betas may be
apprenticing to develop effective and appropriate displays that enhance their subsequent success as alphas. Complex social affiliations
appear to mediate selective pressure for cooperation in this species.
Keywords: lance-tailed manakin, Chiroxiphia, cooperative courtship,
relatedness, delayed direct benefits.
Cooperative reproduction is a classic paradox in evolutionary biology because cooperating individuals apparently
sacrifice their own reproductive potential to assist the reproductive efforts of others. Much theory about cooperation is informed by studies of cooperative breeding, in
* Present address: Max Planck Institute for Ornithology, Department of Behavioural Ecology and Evolutionary Genetics, Postfach 1564, Haus Nr. 5, D82319 Seewiesen, Germany; e-mail: [email protected].
Am. Nat. 2007. Vol. 169, pp. 423–432. 䉷 2007 by The University of Chicago.
0003-0147/2007/16904-41965$15.00. All rights reserved.
which certain individuals delay reproduction to help others
raise offspring (Brown 1987; Cockburn 1998; Koenig and
Dickinson 2004). In many cases, the paradox of cooperation is resolved by inclusive fitness theory (Hamilton
1964): cooperators help close relatives and benefit through
the increased production of nondescendant kin (Griffin
and West 2003). However, cooperators may reap direct
benefits as well (e.g., Richardson et al. 2002; Dickinson
2004). As cooperators most frequently live in groups of
close relatives, it can be difficult to separate the influence
of direct and indirect genetic benefits in selecting for cooperative behavior (Clutton-Brock 2002). Species and situations in which helping is decoupled from the immediate
family group are therefore of particular interest in investigating the evolution of cooperation.
Although animals are also known to form cooperative
alliances in contexts other than helping at the nest, these
have received relatively little empirical attention (Cockburn 1998). Cooperative alliances in animal societies may
form to increase foraging efficiency (Parabuteo unicinctus:
Coulson and Coulson 1995; Phalacrocorax carbo sinensis:
Van Eerden and Voslamber 1995), acquire or defend territories (Psophia leucoptera: Sherman 1995; Alouatta seniculus: Pope 1990; Panthera leo: Grinnell et al. 1995; Heinsohn et al. 1996), or attract or defend mates (Tursiops sp.:
Krützen et al. 2004; Meleagris gallopavo: Watts and Stokes
1971; Krakauer 2005; Philomachus pugnax: van Rhijn 1973;
Lank et al. 2002; Chiroxiphia manakins: Sick 1967). Ecological benefits from cooperative foraging or territoriality
are shared among members of a group; in contrast, the
reproductive benefits of alliances formed to attract mates
are not usually distributed evenly among members of the
coalition (Kokko and Johnstone 1999). Such reproductive
coalitions are therefore ideal for investigating the complex
series of costs and benefits that interact to determine
whether subordinate individuals cooperate.
Manakins in the genus Chiroxiphia are a particularly
interesting example of reproductive coalitions. Male Chiroxiphia form long-term cooperative alliances of a dominant alpha male and one or more subordinate beta partners to perform tightly synchronized courtship dances in
424
The American Naturalist
a dispersed lek mating system (Gilliard 1959; Foster 1977,
1981). A series of previous studies on long-tailed manakins
(Chiroxiphia linearis) have indicated that beta males do
not benefit by helping relatives (McDonald and Potts
1994), and copulations by subordinate males are exceedingly rare (Foster 1981; McDonald 1993). These results
contradict findings in other species with cooperative courtship that suggest that individuals receive inclusive or direct
fitness benefits from cooperative behavior (Lank et al.
2002; Krakauer 2005). Instead, male Chiroxiphia manakins
are believed to benefit by eventually inheriting a breeding
position from their alpha partner (McDonald and Potts
1994). Until now, however, the genetic measures of direct
fitness and experimental assessments of territory inheritance necessary to evaluate definitively the major adaptive
hypotheses about subordinate cooperation have been
lacking.
To investigate the adaptive benefits of male cooperative
courtship, I studied cooperative behavior in the lancetailed manakin (C. lanceolata). I tested the predictions of
three classic, nonexclusive hypotheses regarding the types
of benefits that betas may obtain through their cooperative
behavior: (1) direct genetic benefits: if beta-ranked males
receive direct fitness benefits from mating with the females
they help to attract, then beta copulations will be detected
by genetic paternity testing; (2) inclusive fitness benefits:
if beta males receive inclusive fitness benefits by increasing
relatives’ reproductive success (Hamilton 1964), betas will
be more related to their partners than to other males selected at random from the population; and (3) delayed
direct benefits: if beta males gain future direct fitness benefits from cooperation, betas should become alphas with
greater frequency than do nonbetas. In addition, if delayed
direct benefits result from queuing for status (Wiley and
Rabenold 1984), betas should ascend to alpha status on
the display area where they cooperate.
Similar to other Chiroxiphia manakins, pairs of male
lance-tailed manakins form multiyear partnerships to execute a stereotyped courtship display that includes both a
coordinated duet song and a dance display (DuVal 2007a).
Females move freely among display areas in a dispersed
lek to assess potential mates. Each display area is typically
attended by one alpha, one beta, and on average 10 other
adult- and subadult-plumage males that do not display
for females (DuVal 2007b). Only alphas appear to have
the opportunity to mate because only alphas perform the
solo displays that immediately precede copulation (DuVal
2007b), raising the question of why beta males cooperate.
This study combines behavioral, genetic, and experimental
evidence to assess the adaptive basis of cooperative courtship for subordinate males in lance-tailed manakin
coalitions.
Methods
Field Methods
I studied lance-tailed manakins on 46 ha of Isla Boca
Brava, Chiriquı́, Panamá (8⬚12⬘N, 82⬚12⬘W). Using mist
nets, I captured a total of 457 postfledging individuals
during 1999–2004. All birds were color-banded with
unique band combinations, and approximately 20 mL of
blood were taken from the brachial wing vein and stored
in lysis buffer (DuVal and Nutt 2005). Behavioral observations were conducted in March–July 2000–2002,
March–May 2003, and March–April 2004.
Nests were located by daily searches of understory vegetation. Lance-tailed manakins lay a maximum of two eggs
per clutch (mean 1.88 Ⳳ 0.33 eggs, n p 170 nests). A total
of 218 nestlings from 130 broods were genetically sampled
between 2000 and 2003. The adult female attending a nest
was assigned as the female parent of the chicks in that
nest.
Male lance-tailed manakins displayed on a dispersed lek
consisting of approximately 25–30 display areas spread
throughout the study site. Display areas ranged from 525
to 4,500 m2 in size and were in auditory but not visual
contact (DuVal 2007a). The study site is bordered on two
sides by ocean and on the third by cattle pasture that is
generally unsuitable for lance-tailed manakin display areas,
but remnant forest along the coastline did host additional
male territories that females nesting on the study site may
have sampled. Each display area was attended by multiple
adult and subadult males, though only one alpha male and
usually one beta male displayed for females at each area
(see “Behavioral Definitions”; DuVal 2007b). Individual
behavior at display areas was monitored at 16–28 display
areas per year (4,146 h total observation) during 1- or 2h sessions of all-occurrence sampling (Altmann 1974).
Courtship behaviors included “pip flight” attraction displays performed when females approach the display area
(DuVal 2007a), “leapfrog” displays by two males, and nine
other display elements that are characterized elsewhere
(DuVal 2007a). Observers recorded identities of all birds
seen as well as frequencies and durations of all display
behaviors.
Behavioral Definitions
The alpha and beta partners were the most common duet
singers and the only males on the display area to perform
displays for females. The alpha in each partnership (1) was
present for the greatest proportion of observation sessions,
(2) gave the “eek” call ending a bout of display for females,
(3) performed solo pip flights, and (4) performed solo
slow-flight displays for females. The distinction here of
displays “for females” specifically denotes a female present
Advantages of Cooperative Courtship in Manakins
425
Table 1: Characteristics of dinucleotide microsatellite loci, based on genotypes of 308 postfledging individuals used to calculate
allele frequencies for relatedness analyses
Locus
No.
alleles
Size
rangea
Lan10
Lan15
Lan20
Lan21
Lan22
LTM8
LTM15
Lox1
Man3
5
2
8
6
7
4
2
3
8
192–212
202–206
105–151
198–214
143–165
137–145
174–194
397–409
193–225
Heterozygosityb
HO
HE
Tac
Null
frequencyd
Genotyping
error ratee
Referencef
.689
.134
.721
.725
.543
.521
.387
.351
.767
.710
.137
.748
.742
.509
.537
.448
.352
.752
54
50
54
52
55
54
52
50
50
⫹.017
⫹.008
⫹.018
⫹.008
⫺.048
⫹.014
⫹.072
⫹.006
⫺.016
.025
.004
.010
.010
.009
.008
.039
.009
.015
DuVal and Nutt 2005
DuVal and Nutt 2005
DuVal and Nutt 2005
DuVal and Nutt 2005
DuVal and Nutt 2005
McDonald and Potts 1994
McDonald and Potts 1994
Piertney et al. 1999a
Piertney et al. 2002
a
Size range of alleles detected at each locus, in base pairs.
Observed (HO) and expected (HE) proportion of heterozygotes.
c
Annealing temperature for polymerase chain reaction (PCR).
d
Per-locus frequency of null alleles, as calculated by CERVUS (Marshall et al. 1998). No locus was found to deviate from Hardy-Weinberg expectations.
e
Per-locus genotyping error was estimated from 57 to 108 randomly selected individuals that were repeat-genotyped in independent PCR reactions.
Individuals were typed on average 2.3 Ⳳ 0.98 times. Of 1,848 total PCR reactions, 26 resulted in incorrect allele calls, giving an overall error rate of
0.014 per reaction.
f
Reference in which locus or loci were found.
b
on the display perch in close proximity to displaying males;
similar male behaviors also occurred when no females were
present, but these were not interpreted as status indicators.
Using these criteria, I assigned alpha and beta status to
13–21 pairs per year (85 display area–years). A logistic
regression model for assigning male status, detailed elsewhere, was used to confirm qualitative status assignments
for 2001–2004 (DuVal 2007b). All alpha or beta males (and
the majority of all individuals) sighted after 2001 were
banded. Behavioral data comprised only observations in
which all interacting individuals were identified. In each
year, 30.7 Ⳳ 5.8 males were identified as alphas or betas,
but 73.7 Ⳳ 18.6 additional adult-plumage males were also
present at observed display areas (n p 4 years; DuVal
2007b). To characterize the source of males replacing alpha
and beta males, I defined two further status classes of adult
males: floaters and affiliates. Floater males were adult males
that were sighted at three or more display areas but were
not consistently present in any one area. Affiliates were
adult males that were consistently present at one display
area and performed duet songs and participated in dance
displays when no females were present but that were neither alpha nor beta at that display area.
DNA Extraction and Genotyping
Genomic DNA was extracted from blood using the DNeasy
Tissue Kit (Qiagen). Individuals were genotyped at nine
variable microsatellite loci with a combined paternity exclusion power of 0.98 (table 1). Loci showed no evidence
of linkage disequilibrium, as determined by GENEPOP,
version 3.1d (Raymond and Rousset 1995), using Bon-
ferroni correction of observed P values (a p 0.001). Fluorescent-labeled alleles were amplified using polymerase
chain reaction (PCR) under the following reaction conditions: 5 min at 94⬚C; 32 cycles of 30 s at 94⬚C, 30 s at
Ta (table 1), and 45 sec at 72⬚C; and a final extension of
20 min at 72⬚C. Reactions were 10 mL total volume with
5–20 ng of genomic DNA; 0.5 U Taq DNA polymerase
(Invitrogen); 1 mL Taq DNA polymerase PCR reaction
buffer (10 mM Tris-HCl, 1.5 mM MgCl2, 50 mM KCl, pH
8.3); 0.2 mL each of forward and reverse primers (10 mM)
and 0.2 mL dNTPs (40 mM). PCR products were visualized
on an Applied Biosystems 3730 DNA analyzer.
Analysis of Genetic Data
Genetic estimates of reproductive success were generated
by determining paternity of chicks during 2000–2003. Genotypes of all chicks were compatible with their behaviorally assigned mother (see “Field Methods”). To avoid
assigning paternity to males that matched offspring by
chance (Type I error), paternity was assigned through a
combination of maximum likelihood analysis (CERVUS,
ver. 2.0; Marshall et al. 1998) and genotypic exclusion. In
this approach, a male was assigned as a chick’s father only
if he matched a chick with ≥95% confidence and was the
only perfect genotypic match among the candidate males
with positive likelihood scores. Confidence levels for paternity assignment are calculated in CERVUS via a simulation employing population allele frequencies, the number of candidate males, and sampling success. This
simulation returns a “delta score” used to resolve the likelihood of paternity between candidate males at a given
426
The American Naturalist
confidence level. All simulations were conducted using
10,000 cycles and a 1% genotyping error based on the
empirical per-locus genotyping error (table 1). Yearspecific allele frequencies and proportion of loci typed
(0.98 in all years) were calculated from the genotypes of
postfledging individuals that were captured or resighted
during the year in question. Separate paternity analyses
were performed for each study year, with all alpha and
beta males present on the study area designated as candidate fathers. In cases in which one alpha performed
courtship displays and duets with two adult males at approximately equal rates, both males were treated as beta
partners and as candidate parents in paternity analyses.
The proportion of candidates sampled was defined as the
number of identified alpha or beta males divided by two
times the number of display areas for which behavioral
observations were performed (table 2). Not every display
area active on the study site was observed in every year,
so this is likely an overestimate of candidate male sampling
success. Decreasing this sampling parameter decreased the
number of chicks assigned to fathers but did not change
the relative success of alpha versus beta males.
Paternity analyses assigned 63 of 218 tested chicks (representing 52 of 130 broods) to a known-status male (table
2). Four additional chicks were assigned to three males
known to be either alpha or beta at observed display areas
but whose exact status was undetermined; these chicks
were excluded from tests of alpha versus beta reproductive
success. Remaining chicks were unassigned for one of three
reasons: mismatches with all males (which may represent
mutations, genotyping error, or failure to sample the true
father; n p 42); complete genotypic matches with multiple candidate males (n p 87); or low statistical power
to assign paternity (n p 22). While genotyping at additional loci may resolve paternity of these chicks, there was
no indication that chicks that were assigned paternity were
not a random, representative sample of the young
produced.
Pairwise relatedness (r) between alpha and beta partners
was calculated using the program Relatedness, version 5.0
(Queller and Goodnight 1989), with background allele frequencies estimated from 308 postfledging individuals. The
r values of 58 known alpha-beta pairs were compared with
those of 58 artificial pairs drawn at random from observed
alpha and beta males. To validate r values for known pairs
of close kin, I calculated mother-offspring relatedness using one randomly selected chick for each sampled mother
and nestmate relatedness using all two-chick broods.
Alpha Male Removal Experiment
Alpha males were mist-netted and removed from eight
randomly selected display areas (four each in 2002 and
2003, removals separated by ≥10 days). Removed males
were relocated to other islands (n p 5) or collected under
license as museum specimens (n p 3). The two relocated
males transferred to the nearest sites (0.75 and 1 km from
the study site) returned to their territories (after 17 and
124 days, respectively) and were excluded from analysis
of long-term results. Males transferred to more distant sites
(15.25 km from the study site) were not resighted and
thus were considered permanently removed. Observations
were conducted for 2–6 h/day for the 10 days following
each removal to determine when a new male assumed
alpha status. To determine the short-term effect of alpha
removal on beta behavior, I analyzed the proportion of 10
observation sessions in which each beta performed statuslinked behaviors (DuVal 2007b) before (≤29 days) and
after (≤10 days) alpha removal. Proportions were arcsine
transformed and compared in a linear mixed effects model
with year and display area as random effects in the program R (Venables and Ripley 2002). Because the outcomes
of natural turnovers were usually not detected until the
following breeding season, I continued to monitor experimental territories in the following field season (10–12
months after manipulation).
Table 2: Summary of paternity analyses by year: CERVUS input parameters and paternity assignment success
Reasons that chicks were not assigned paternitya
Analysis parameters
2000
2001
2002
2003
Totals
a
No.
candidate
males
Proportion
candidates
sampled
40
38
56
60
.83
.97
.95
.95
95%D
No. chicks
assigned
paternity
Low confidence,
one match
Mismatch most
likely male
Match multiple
males
2.17
1.36
1.65
1.67
10
15
25
17
6
4
7
5
13
13
9
7
10
23
23
31
67
22
42
87
“Match” indicates that all paternal alleles in an offspring’s genotype (determined by subtracting the known maternal alleles) were also
identified in the candidate father’s genotype. Exceptions to this at one or more loci were considered “mismatches.”
Advantages of Cooperative Courtship in Manakins
427
Results
Current Direct Fitness
If current direct fitness selects for cooperation in lancetailed manakins, then subordinate males must reproduce
during their beta tenure. However, 100% of observed copulations (n p 34) were performed by alpha males. Thirtynine percent (17/44) of alpha males were observed copulating with females at display perches versus none of 52
betas (x 2 p 24.4, df p 1, P ! .001).
Since observed copulations may not reflect actual mating success (Hughes 1998), I also evaluated beta reproductive success using genetic analyses of paternity. Genetic
data showed that beta males did sometimes engage in successful copulations (fig. 1). The demonstration here of
successful fertilization by subordinate male lance-tailed
manakins confirms that rare copulations may be an important source of direct fitness for some betas. However,
very few beta males sired chicks (!4%, n p 2 of 52 betas),
suggesting that other adaptive benefits are more important
in selecting for cooperation in this species.
Indirect Fitness
The indirect fitness hypothesis asserts that beta males
partner with closely related alphas, thereby benefiting
from cooperation via the increased production of nondescendant kin. This predicts that betas will be more
closely related to their partners than expected from random pairing in the population. Contrary to this, dyads
of alpha and beta males displaying together were not
more closely related than randomly created pairs of males
(fig. 2). Observed alpha and beta male dyads were related
at r p ⫺0.01 Ⳳ 0.28 (95% confidence interval [CI] p
⫺0.08 to 0.06; n p 58), a level not significantly different
from relatedness between randomly drawn males (r p
0.00 Ⳳ 0.27, 95% CI p ⫺0.07 to 0.07, n p 58; twotailed t-test, t p ⫺0.16, df p 1, 114, P p .87). Relatedness values of mother-chick and of nestmate pairs agree
with values predicted by their pedigree relationships
(mother-offspring, r p 0.49 Ⳳ 0.17, 95% CI p 0.45 to
0.53, n p 84; nestmate, r p 0.45 Ⳳ 0.27, 95% CI p
0.40 to 0.51, n p 86; two-tailed t p 1.05, df p 1, 168,
P p .29), confirming that low r values between alphas
and betas reflect a lack of genetic kinship.
Relatedness could nevertheless play a role in selecting
for beta cooperation if partnerships composed of more
closely related males are more successful in attracting
females than are partnerships composed of unrelated
males. However, male pairs in which alphas sired chicks
were not more closely related than pairs for which no
paternity was detected (successful pair, r p 0.00 Ⳳ 0.26,
95% CI p ⫺0.13 to 0.13, n p 20; unsuccessful pair,
Figure 1: Genetically determined reproductive success of males in different status classes. Of 63 chicks for which paternity could be assigned,
all but two (97%) were sired by alpha males (x2 p 39.8, df p 1, P !
.001; A). Based on the assigned offspring, 49% of alpha males sired chicks
in 2000–2003 versus only 4% of betas (B). Significantly more alpha than
beta individuals sired chicks (x2 p 22.3, df p 1, P ! .001). These data
represent 52 broods by 42 females; chicks sired by betas were from different mothers.
r p ⫺0.02 Ⳳ 0.30, 95% CI p ⫺0.11 to 0.08, n p 38;
two-tailed t p ⫺0.20, df p 56, P p .84; means Ⳳ SD
throughout).
Collectively, these data fail to provide evidence that indirect fitness benefits influence male cooperative courtship
in lance-tailed manakins. These results are similar to findings for the long-tailed manakin (McDonald and Potts
1994), suggesting that cooperation in the genus Chiroxiphia cannot be explained by indirect fitness benefits.
428
The American Naturalist
disappeared in the following breeding season (table 3).
Males replacing alphas came from a variety of sources and
were not limited to the display area where the replacement
occurred (table 4). At the majority of display areas where
an alpha or beta was replaced, new males had no known
history of affiliation at the display area where the vacancy
occurred (58% of 33 replacement events), suggesting that
males do not form strictly linear queues for status within
display areas. However, queuing may still be an important
mechanism for ascending to alpha status, as six (60%) of
10 betas that took over alpha positions did so through
succession. Furthermore, local betas are clearly more likely
than a randomly chosen male to succeed their alpha partners: the probability of randomly selecting the local beta
from the minimum of 80 adult-plumage males present in
each year is 0.013, whereas the observed probability of
local beta succession (0.46) is 35 times this value.
Alpha Removal Experiment
Figure 2: Mean relatedness values of observed alpha-beta partners. Dotted lines indicate expected average r for full siblings or parent-offspring
comparisons (0.5) and unrelated individuals (0). Bars indicate standard
errors. Common letters denote groups that are not statistically different
in relatedness but are significantly different from groups marked by different letters (two-tailed t-test: alpha-beta vs. nestmate t p ⫺10.03,
df p 1, 142; alpha-beta vs. mother-chick t p ⫺13.13, df p 1, 140; both
P ! .001).
Delayed Direct Fitness
The critical prediction of the delayed direct fitness hypothesis is that beta males become alphas more frequently
than do other, nonbeta adult males. Observations of male
status over years support this prediction; of all adult males
observed in one year and resighted in the following year,
betas became alphas in 10 of 67 cases, while nonbeta adults
became alpha in only 6 of 164 cases (15% vs. 4%, x 2 p
6.94, df p 1, P p .008).
This finding raises questions regarding the mechanisms
that provide cooperating betas with increased opportunities to attain alpha status. Queuing for dominance within
a social group is a common mechanism of attaining status
in a variety of species (Alberts et al. 2003; Cockburn et
al. 2003; Buston 2004), and strict linear queues apparently
regulate status succession in the long-tailed manakin
(McDonald and Potts 1994). If queuing for alpha status
is the primary benefit of cooperative behavior, beta lancetailed manakins should consistently ascend to dominant
status when their alpha partner is lost. To test this prediction, I monitored natural territory turnovers by determining male status at display areas in consecutive breeding
seasons. The beta male succeeded his alpha partner in only
six (46%) of 13 cases in which an identified alpha male
To explore the extent of queuing for status experimentally,
I randomly selected eight display areas (four each in 2002
and 2003) and removed the alpha males from these sites.
During the 10 days following the alpha removals, seven
of eight beta males (88%) began to exhibit alpha-specific
behaviors (fig. 3). Thus, immediately following removals,
beta males were significantly more likely to inherit alpha
status than expected given the proportion of beta males
that inherit following natural losses of alphas (beta inheritance after experimental manipulation p 0.88, expected beta inheritance from natural status changes p
0.46; G(1) p 6.03, P ! .02). To determine whether experimentally promoted betas retained alpha positions over a
biologically significant timescale, I revisited six of eight
manipulated display areas during the breeding season following the alpha removal (see “Methods”). Of four experimentally promoted beta males that were resighted during the following breeding season, only one had retained
alpha status. The remaining three males were each beta to
a new alpha partner, either at different display area
(n p 2) or on the original manipulated territory (n p
1). This result strongly suggests that betas that initially
took over vacated alpha positions did not typically maintain this status long enough to gain substantive direct fitness benefits.
Discussion
This study suggests that delayed direct fitness is the primary adaptive advantage attained through cooperation by
beta male lance-tailed manakins. Both behavioral and genetic evidence confirmed that beta males had a low probability of reproducing during their beta tenure, so current
Advantages of Cooperative Courtship in Manakins
429
Table 3: Details of natural transitions in dominance status by display area
Changes in alpha or beta individuals at display
areas
No. display areas
alpha and beta
identified in consecutive yearsa
No change
Both
replaced
Alpha only
replacedb
Beta only
replacedb
2000–2001
2001–2002
2002–2003
2003–2004
11
16
18
14
7
5
12
7
2
3
2
0
0
2 (1)
1
2
2
6 (3)
3 (1)
5
Totals
59
31
7
6c
Years
20d
a
Natural changes in alpha and beta status occurred primarily between field seasons (85%, n p 33
changes), and so male status at a given display area had to be assigned in two consecutive years to identify
partnership changes. Display areas manipulated in the removal experiment were excluded from this analysis
in the season following manipulation but included in subsequent years.
b
Numbers in parentheses represent changes that occurred during a field season in the second of the
two consecutive years.
c
A majority of replaced alphas (10 of 13) were never seen again, and these males were presumed
dead. Three alphas were resighted following their replacement; these individuals were never seen more
than twice and were therefore classified as being of unknown status.
d
Nine of the replaced betas were resighted after they were replaced. Eight moved to a different display
area and were identified as alpha (four males), beta (three males), or present as a nondisplaying male
(one male). One replaced beta remained at the original display area as a nondisplaying male. Two of four
betas that left one display area to become alpha at another had been affiliated simultaneously with both
involved display areas in the year before the partnership change.
direct benefits had a relatively small potential to select for
cooperation in this species. Male partners were not more
closely related than randomly selected pairs of males, and
therefore, betas did not receive indirect fitness benefits
from their cooperative behavior. This finding is particularly important in light of recent evidence that relatedness
influences cooperative courtship (Krakauer 2005) and is
related to spatial associations of several lekking birds (Höglund et al. 1999; Petrie et al. 1999; Piertney et al. 1999b;
Shorey 2002; but see also Gibson et al. 2005). Betas did,
however, become alphas more often than nonbeta males,
providing a clear benefit to cooperative behavior.
This study confirms previous findings that cooperating
partners in Chiroxiphia manakins are unrelated (McDonald and Potts 1994) and that factors selecting for cooperation in this genus are distinct from those detected
in other species with cooperative courtship displays. Subordinate males in coalitions of wild turkeys cooperate with
close relatives and benefit through an increase in indirect
fitness (Krakauer 2005), while male ruffs of both cooperating morphs mate with females that they attract with
paired displays (Lank et al. 2002). There are, therefore,
three emerging pathways for male-male cooperation, each
with its own balance of costs and benefits. Courtship behavior in ruffs, wild turkeys, and Chiroxiphia manakins
apparently represents a case of convergence in which different selective environments result in similar patterns of
cooperative male behavior.
In addition to the adaptive benefits addressed here, co-
operating individuals may benefit through reciprocal altruism (Trivers 1971), in which the helper receives a direct
benefit from help in kind by the recipient of his altruistic
behavior. Reciprocal altruism cannot apply in this system,
as alpha and beta partners never switched behavioral roles.
However, cooperation by beta male lance-tailed manakins
may be viewed as a form of psuedoreciprocity (Connor
1995) or “stakeholder altruism” (Roberts 2005), in which
helpers receive future by-product benefits from their beneficiary’s selfish behavior. In the lance-tailed manakin sysTable 4: Source of males replacing alphas and betas after
natural transitions
Prior display area
same or different
from where replacement
occurred?
Alpha
Beta
Affiliatea
Alpha
Beta
Affiliate
Floater
Unknowna
Total
a
Same
Same
Same
Different
Different
Different
Not applicable
Not applicable
New
alphas
New
betas
Total
…
6
0
2
1
0
2
2
0
…
8
0
2
5
7
11
0
6
8
2
3
5
9
13
13
33
46
Unknown males were not observed more than five times between
banding and assuming alpha or beta status.
430
The American Naturalist
Figure 3: Behavioral changes by beta males in the 10 days following
experimental removal of their alpha partners. Bars indicate the proportion
of observation sessions in which betas performed behaviors before (white)
and after (shaded) the removal of their alpha partner (n p 8 beta males).
Horizontal lines indicate median values, boxes surround the 25%–75%
intervals of the data, and vertical lines show data within 1.5 interquartile
ranges of this interval. Betas performed only solo pip flight (attraction)
displays after their partners were removed (linear mixed effects [LME]:
t p ⫺3.12, df p 7, P p .02). Several betas also performed solo displays
for females on the dance perch, but this increase in display rate was not
significant (LME: t p 1.48, df p 7, P p .18). In general, betas continued
to perform duet songs and were present for the majority of observation
sessions following the removal of their alpha partner (LME: t p 0.14 for
duets and ⫺0.67 for presence, df p 7 , P 1 .5 for both). While some males
also formed new partnerships and performed two-male displays for females on the dance perch (n p 3 former betas), the overall rate of paired
displays decreased when compared with display rate when the removed
alpha was present (LME: t p 2.72, df p 7, P p .03).
tem, this would be the case if the alpha’s current reproductive success increases his beta partner’s later
reproductive success. Female fidelity to display sites provides one mechanism by which this might occur; in longtailed manakins, females return to the same display sites
to copulate, even when the attending alpha male is replaced
(McDonald and Potts 1994). However, such a mechanism
would predict high site fidelity by beta males as well, which
was not the case in this study. Pseudoreciprocity could also
result if betas benefit from interactions with alphas by
learning skills involved in successful courtship displays. By
helping alphas attract females, betas witness and participate in more successful courtship displays, which may improve their own display competence when they later become alphas. More comprehensive analyses of male
reproductive success and female mate choice in multiple
years offer one means of testing these possibilities.
Importantly, both natural and experimental changes in
male status indicate that territory inheritance does not
occur in a linear queue, as previously suggested for the
congeneric long-tailed manakin (Foster 1977). Though
queuing cannot be ruled out as a mechanism of attaining
alpha status in lance-tailed manakins because local betas
were more likely than random males to succeed their alpha
partners, the observed results suggest that territory inheritance is modified by factors in addition to male status
and presence. Departure from a queuing system could be
explained if some betas are unable to maintain alpha status
during prolonged interactions with other, competitively
superior males. Alternatively, there are three mechanisms
by which beta males may obtain benefits from associating
with alphas in the absence of a strict queuing system. First,
cooperation may provide opportunities to assess territory
quality (Boulinier et al. 1996; Hatchwell et al. 1999). However, adult males that do not display for females are frequently observed at display areas (e.g., 38.3% Ⳳ 15.4% of
observation sessions in 2002, n p 22 display areas) and
are tolerated by the alpha and beta males, suggesting that
cooperation in displays is not necessary to gather information about territory quality. Second, cooperative interactions with a variety of males may help betas develop
affiliations with a future subordinate partner. However,
beta males affiliate almost exclusively with their alpha partner for cooperative duets and dance displays (DuVal
2007b), thereby limiting the time that betas can spend
developing affiliations with other males. Third, cooperation may increase skill in display performance (Trainer et
al. 2002). Lance-tailed manakin courtship displays are long
and complex (DuVal 2007a), and interactions with experienced males may be a critical component of learning
display behavior (Collis and Borgia 1993). In accord with
this hypothesis, beta males are generally younger than their
alpha partners (DuVal 2007b). Consistent performance of
courtship displays with a successful alpha partner may
allow betas to develop effective and appropriate displays
that enhance their subsequent success as alphas. In systems
such as this, in which factors other than kinship select for
complicated cooperative behavior, long-term strategies to
maximize future fitness may depend on social affiliations
that reinforce the evolution of complex social structure.
Acknowledgments
I thank J. Dale, J. Dickinson, R. Gibson, M. Hauber, B.
Kempenaers, W. Koenig, A. Krakauer, P. Palsbøll, S. PruettJones, and especially E. A. Lacey for helpful comments on
this manuscript. D. McDonald offered advice and encouragement in the early stages of this research. Field assistance was provided B. Carter, K. Janaes, R. Lorenz, J.
Lorion, K. Manno, E. Reeder, and M. Westbrock. Frank
and Y. Köhler kindly allowed field site access. Funding for
this research was provided by the National Science Foundation (predoctoral fellowship and dissertation improvement grant); the University of California, Berkeley, Mu-
Advantages of Cooperative Courtship in Manakins
seum of Vertebrate Zoology, Department of Integrative
Biology, and Vice Chancellor for Research fund; Animal
Behavior Society; American Ornithologists’ Union; American Museum of Natural History; Sigma Delta Epsilon
Graduate Women in Science; Manomet Bird Observatory
Kathleen Anderson award; Sigma Xi; and the Smithsonian
Tropical Research Institute. All protocols followed Animal
Behavior Society guidelines for research on animals and
were approved by the University of California, Berkeley,
Animal Care and Use Committee and Autoridad Nacional
del Ambiente, Panamá.
Literature Cited
Alberts, S. C., H. E. Watts, and J. Altmann. 2003. Queuing and queuejumping: long-term patterns of reproductive skew in male savannah baboons, Papio cynocephalus. Animal Behaviour 65:821–840.
Altmann, J. 1974. Observational study of behavior: sampling methods. Behaviour 48:227–265.
Boulinier, T., E. Danchin, J. Y. Monnat, C. Doutrelant, and B. Cadiou.
1996. Timing of prospecting and the value of information in a
colonial breeding bird. Journal of Avian Biology 27:252–256.
Brown, J. L. 1987. Helping and communal breeding in birds: ecology
and evolution. Princeton University Press, Princeton, NJ.
Buston, P. M. 2004. Territory inheritance in clownfish. Proceedings
of the Royal Society B: Biological Sciences 271(suppl.):252–254.
Clutton-Brock, T. 2002. Breeding together: kin selection and mutualism in cooperative vertebrates. Science 296:69–72.
Cockburn, A. 1998. Evolution of helping behavior in cooperatively
breeding birds. Annual Review of Ecology and Systematics 29:141–
177.
Cockburn, A., H. L. Osmond, R. A. Mulder, D. J. Green, and M. C.
Double. 2003. Divorce, dispersal and incest avoidance in the cooperatively breeding superb fairy-wren Malurus cyaneus. Journal
of Animal Ecology 72:189–202.
Collis, K., and G. Borgia. 1993. The costs of male display and delayed
plumage maturation in the satin bowerbird (Ptilonorhynchus violaceus). Ethology 94:59–71.
Connor, R. C. 1995. Altruism among non-relatives: alternatives to
the “Prisoner’s Dilemma.” Trends in Ecology & Evolution 10:84–
86.
Coulson, J. O., and T. D. Coulson. 1995. Group hunting by Harris’
hawks in Texas. Journal of Raptor Research 29:265–267.
Dickinson, J. L. 2004. A test of the importance of direct and indirect
fitness benefits for helping decisions in western bluebirds. Behavioral Ecology 15:233–238.
DuVal, E. H. 2007a. Cooperative display and lekking behavior of the
lance-tailed manakin (Chiroxiphia lanceolata). Auk (forthcoming).
———. 2007b. Social organization and variation in cooperative alliances among male lance-tailed manakins. Animal Behaviour, doi:
10.1016/j.anbehav.2006.05.017.
DuVal, E. H., and K. J. Nutt. 2005. Isolation and characterization of
polymorphic microsatellite loci in the lance-tailed manakin (Chiroxiphia lanceolata). Molecular Ecology Notes 5:112.
Foster, M. S. 1977. Odd couples in manakins: a study of social organization and cooperative breeding in Chiroxiphia linearis. American Naturalist 111:845–853.
———. 1981. Cooperative behavior and social organization of the
431
swallow-tailed manakin (Chiroxiphia caudata). Behavioral Ecology
and Sociobiology 9:167–177.
Gibson, R., D. Pires, K. Delaney, and R. Wayne. 2005. Microsatellite
DNA analysis shows that greater sage grouse leks are not kin
groups. Molecular Ecology 14:4453–4459.
Gilliard, E. T. 1959. Notes on the courtship behavior of the bluebacked manakin (Chiroxiphia pareola). American Museum Novitates 1942:1–19.
Griffin, A. S., and S. A. West. 2003. Kin discrimination and the benefit
of helping in cooperatively breeding vertebrates. Science 302:634–
636.
Grinnell, J., C. Packer, and A. E. Pusey. 1995. Cooperation in male
lions: kinship, reciprocity or mutualism? Animal Behaviour 49:
95–105.
Hamilton, W. D. 1964. The genetical evolution of social behaviour.
I, II. Journal of Theoretical Biology 7:1–52.
Hatchwell, B. J., A. F. Russell, M. K. Fowlie, and D. J. Ross. 1999.
Reproductive success and nest-site selection in a cooperative
breeder: effect of experience and a direct benefit of helping. Auk
116:355–363.
Heinsohn, R., C. Packer, and A. E. Pusey. 1996. Development of
cooperative territoriality in juvenile lions. Proceedings of the Royal
Society B: Biological Sciences 263:475–479.
Höglund, J., R. V. Alatalo, A. Lundberg, P. T. Rintamaki, and J.
Lindell. 1999. Microsatellite markers reveal the potential for kin
selection on black grouse leks. Proceedings of the Royal Society
B: Biological Sciences 266:813–816.
Hughes, C. 1998. Integrating molecular techniques with field methods in studies of social behavior: a revolution results. Ecology 79:
383–399.
Koenig, W. D., and J. L. Dickinson. 2004. Ecology and evolution of
cooperative breeding in birds. Cambridge University Press, New
York.
Kokko, H., and R. A. Johnstone. 1999. Social queuing in animal
societies: a dynamic model of reproductive skew. Proceedings of
the Royal Society B: Biological Sciences 266:571–578.
Krakauer, A. 2005. Kin selection and cooperative courtship in wild
turkeys. Nature 434:69–72.
Krützen, M., L. M. Barré, R. C. Connor, J. Mann, and W. B. Sherwin.
2004. “O father: where art thou?”—paternity assessment in an
open fission-fusion society of wild bottlenose dolphins (Tursiops
sp.) in Shark Bay, Western Australia. Molecular Ecology 13:1975–
1990.
Lank, D. B., C. M. Smith, O. Hanotte, A. Ohtonen, S. Bailey, and
T. Burke. 2002. High frequency of polyandry in a lek mating system. Behavioral Ecology 13:209–215.
Marshall, T. C., J. Slate, L. E. B. Kruuk, and J. M. Pemberton. 1998.
Statistical confidence for likelihood-based paternity inference in
natural populations. Molecular Ecology 7:639–655.
McDonald, D. B. 1993. Demographic consequences of sexual selection in the long-tailed manakin. Behavioral Ecology 4:297–309.
McDonald, D. B., and W. K. Potts. 1994. Cooperative display and
relatedness among males in a lek-mating bird. Science 266:1030–
1032.
Petrie, M., A. Krupa, and T. Burke. 1999. Peacocks lek with relatives
even in the absence of social and environmental cues. Nature 401:
155–157.
Piertney, S. B., M. Marquiss, and R. Summers. 1999a. Characterization of tetranucleotide microsatellites in the Scottish crossbill
(Loxia scotica). Molecular Ecology 7:1247–1263.
432
The American Naturalist
Piertney, S. B., A. D. C. MacColl, X. Lambin, R. Moss, and J. F.
Dallas. 1999b. Spatial distribution of genetic relatedness in a moorland population of red grouse (Lagopus lagopus scoticus). Biological
Journal of the Linnean Society 68:317–331.
Piertney, S. B., L. Shorey, and J. Höglund. 2002. Characterization of
microsatellite DNA markers in the white-bearded manakin (Manacus manacus). Molecular Ecology Notes 2:504–505.
Pope, T. R. 1990. The reproductive consequences of male cooperation
in the red howler monkey: paternity exclusion in multi-male and
single-male troops using genetic markers. Behavioral Ecology and
Sociobiology 27:439–446.
Queller, D. C., and K. F. Goodnight. 1989. Estimating relatedness
using genetic markers. Evolution 43:258–275.
Raymond, M., and F. Rousset. 1995. GENEPOP (version 1.2): population genetics software for exact tests and ecumenicism. Journal
of Heredity 86:248–249.
Richardson, D. S., T. Burke, and J. Komdeur. 2002. Direct benefits
and the evolution of female-biased cooperative breeding in Seychelles warblers. Evolution 56:2313–2321.
Roberts, G. 2005. Cooperation through interdependence. Animal Behaviour 70:901–908.
Sherman, P. T. 1995. Social organization of cooperatively polyandrous
white-winged trumpeters (Psophia leucoptera). Auk 112:296–309.
Shorey, L. 2002. Genetic structuring and the evolution of lekking
behaviours in the white-bearded manakin, Manacus manacus. PhD
diss. thesis. Uppsala University.
Sick, H. 1967. Courtship behavior in manakins (Pipridae): a review.
Living Bird 6:5–22.
Trainer, J. M., D. B. McDonald, and W. A. Learn. 2002. The development of coordinated singing in cooperatively displaying longtailed manakins. Behavioral Ecology 13:65–69.
Trivers, R. L. 1971. The evolution of reciprocal altruism. Quarterly
Review of Biology 46:35–57.
Van Eerden, M. R., and B. Voslamber. 1995. Mass fishing by cormorants Phalacrocorax carbo sinensis at Lake Ijsselmeer, The Netherlands: a recent and successful adaptation to a turbid environment. Ardea 83:199–212.
van Rhijn, J. G. 1973. Behavioral dimorphism in male ruffs, Philomachus pugnax (L.). Behaviour 47:153–229.
Venables, W. N., and B. D. Ripley. 2002. Modern applied statistics
with S-plus. Springer, New York.
Watts, C. R., and A. W. Stokes. 1971. The social order of turkeys.
Scientific American 224:112–118.
Wiley, R. H., and K. N. Rabenold. 1984. The evolution of cooperative
breeding by delayed reciprocity and queuing for favorable social
positions. Evolution 38:609–621.
Associate Editor: Elizabeth Adkins-Regan
Editor: Michael C. Whitlock